636. Plasma and cerebrospinal fluid neurofilament light chain in the differential diagnosis of neurodegenerative and primary psychiatric disorders: a systematic review

Abstract

Abstract Background Distinguishing neurodegenerative disorders from primary psychiatric conditions remains a major clinical challenge, particularly in early or atypical presentations. Neurodegenerative disorders may present with prominent psychiatric symptoms, leading to substantial clinical overlap with primary psychiatric conditions and frequent diagnostic delay. Reliance on symptom-based assessment alone is therefore insufficient in clinically ambiguous cases, underscoring the need for objective biomarkers to support differential diagnosis. Neurofilament light chain (NfL) is a well-established marker of neuroaxonal injury and neurodegeneration. Elevated NfL concentrations have been consistently reported across a range of neurodegenerative disorders, whereas levels in primary psychiatric conditions generally remain low. This contrast supports the potential role of NfL in the differential diagnosis between neurodegenerative and psychiatric disorders. Advances in analytical sensitivity have enabled reliable measurement of NfL in both cerebrospinal fluid (CSF) and blood, increasing interest in its diagnostic utility across clinical settings. However, the relative performance and clinical applicability of CSF and plasma NfL for distinguishing neurodegenerative from primary psychiatric disorders remain incompletely defined. Aims & Objectives This systematic review aims to summarise and critically evaluate the evidence supporting the use of NfL measured in CSF and plasma for the differential diagnosis between neurodegenerative disorders and primary psychiatric disorders. CSF and plasma NfL are analysed separately to assess their respective diagnostic performance and potential clinical utility. Method A systematic literature search was conducted in PubMed, supplemented by manual screening of reference lists. Eligible studies involved human participants, quantified NfL in CSF and/or blood, and directly compared neurodegenerative and primary psychiatric disorders. Data extraction focused on study design, sample size, diagnostic groups, biological sample, analytical methods, and reported measures of diagnostic accuracy. Results Thirty-one studies met the inclusion criteria. Of these, twenty-one assessed NfL exclusively in CSF, eight evaluated blood-based NfL alone, and two measured NfL in both CSF and blood. Across studies, NfL concentrations were higher in individuals with neurodegenerative disorders than in those with primary psychiatric disorders. In contrast, psychiatric populations generally exhibited NfL concentrations comparable to healthy control groups. This distinction was observed for both CSF and blood measurements, with limited overlap between diagnostic groups in most cohorts. CSF NfL demonstrated robust and consistent discriminatory capacity across studies, with diagnostic performance frequently described as high. Blood-based NfL also showed clinically meaningful differentiation between neurodegenerative and psychiatric conditions, although greater variability was reported across cohorts. Differences in age distribution, disease stage and analytical methods appeared to influence diagnostic performance, particularly for blood measurements. Several studies reported longitudinal findings, indicating that elevated baseline NfL was associated with subsequent diagnostic reclassification from a primary psychiatric diagnosis to a neurodegenerative disorder, suggesting added value beyond initial clinical assessment. Discussion & Conclusions Overall, the evidence indicates that NfL is a valuable biomarker for distinguishing neurodegenerative disorders from primary psychiatric conditions. CSF NfL shows consistently high diagnostic accuracy and appears particularly useful in diagnostically challenging presentations, while plasma NfL represents a promising, less invasive alternative with broader clinical accessibility. However, heterogeneity limits standardisation and widespread clinical implementation. Future research should prioritise large prospective cohorts to support integration of NfL measurements into diagnostic pathways.